2f5fdd200
Refs #314. Fixed clang compiling bug on OSX. by
2013-11-07 08:12:03 +0800
80a2e901b
added dgemm_tcopy_4_vfpv3.S and sgemm_tcopy_4_vfpv3.S by
2013-11-06 20:01:18 +0100
73770e60b
Refs #309. Fixed trtri_U single thread computational bug. by
2013-11-07 01:08:39 +0800
ac50bccbd
added cgemm_ncopy_2_vfpv3.S and made assembler labels unique by
2013-11-05 20:21:35 +0100
82015beae
added zgemm_ncopy_2_vfpv3.S and made assembler labels unique by
2013-11-05 19:31:22 +0100
6216ab8a7
removed obsolete gemm_kernels from haswell branch by
2013-11-04 08:33:04 +0100
370e3834a
added missing file kernel/arm/Makefile by
2013-11-03 11:54:39 +0100
95aedfa0f
added missing file arm/Makefile in lapack/laswp by
2013-11-03 11:19:32 +0100
cba97daf3
added missing file cblas_noconst.h to the armv7 branch by
2013-11-03 11:04:16 +0100
5400a9f4e
redefined functions for TIMING and YIELDING for ARMV7 processor by
2013-11-03 10:34:04 +0100
e31186efd
deleted obsolete dgemm_kernel and dtrmm_kernel by
2013-11-02 13:12:21 +0100
2b801a00a
small optimizations on sgemm_kernel for ARMV7 by
2013-11-02 13:06:11 +0100
b3eab8fcb
minor optimizations on zgemm_kernel for ARMV7 by
2013-11-02 09:43:53 +0100
6d9d70c55
Fixed #315. Added OPENBLAS_ prefix to openblas_config.h. by
2013-11-02 15:59:00 +0800
dfd1064d7
refs #287. Don't enable OpenMP for netlib LAPACK sequential Fortran codes. by
2013-11-02 15:09:33 +0800
02bc36ac7
added sgemm_ncopy routine and made some improvements on cgemm_kernel for ARMV7 by
2013-11-01 18:22:27 +0100
5118a7f4d
small optimizations on dgemm_kernel for Piledriver by
2013-10-31 11:53:26 +0100
e172b70ea
added cgemm_kernel for Piledriver by
2013-10-31 08:38:17 +0100
1cf4b974b
added zgemm_kernel for Piledriver by
2013-10-30 09:12:17 +0100
7bccff151
added sgemm_kernel for PILEDRIVER by
2013-10-29 22:53:04 +0100
afe44b024
tests and code cleanup of gemm_kernels for HASWELL by
2013-10-28 14:23:48 +0100
a77c71eaf
added highly optimized dgemm_kernel for HASWELL by
2013-10-28 10:23:47 +0100
b2219b347
Merge pull request #311 from loladiro/patch-1 by
2013-10-24 23:41:22 -0700
f5a0038ba
(refs/pull/311/head)
Use FC instead of CC to link the dynamic library on OS X by
2013-10-23 18:43:00 -0400
c93709012
Added gfortran dependency for LSB/lsbcc. by
2013-10-22 13:24:47 +0800
fe8c5666f
optimized dgemm_kernel for HASWELL by
2013-10-20 16:52:26 +0200
f6b50057e
corrected and testet FMA3 Code by
2013-10-19 10:52:20 +0200
2840d56ae
added dgemm_kernel for Piledriver by
2013-10-19 09:47:15 +0200
2d49db2f5
moved compiler flags from Makefile.rule to Makefile.arm by
2013-10-16 19:04:42 +0200
04391e6d9
optimized param.h by
2013-10-16 18:04:34 +0200
85484a42d
added kernels for cgemm, ctrmm, zgemm and ztrmm by
2013-10-16 18:00:41 +0200
3983011f0
added sgemm- and strmm_kernel by
2013-10-14 08:22:27 +0200
2a1515c9d
added dgemm_ncopy_4_vfpv3.S by
2013-10-12 16:48:29 +0200
31f51e78b
minor optimizations on dgemm_kernel by
2013-10-12 09:42:18 +0200
beffee7d9
Fixed buffer overflow bug in kernel/x86_64/dgemv_t.S file. by
2013-10-11 03:20:20 +0800
a35f4343f
Merge pull request #301 from yieldthought/develop by
2013-10-09 00:46:49 -0700
43457f552
(refs/pull/301/merge)
Merge ce5626a384 into 16eb780e13 by
2013-10-08 07:38:28 -0700
ce5626a38
(refs/pull/301/head)
Remove -Wl,--retain-symbols-file from dynamic library linking to fix tool support by
2013-10-08 16:37:17 +0200
e0b968c3a
Changed kernels for dgemm and dtrmm by
2013-10-05 12:59:44 +0200
93f1074dd
changed some values for arm by
2013-09-30 18:03:56 +0200
1c63180bb
updated dgemm_kernel_8x2_vfpv3.S by
2013-09-30 17:31:23 +0200
22a8fcc4b
add modified c_check perl program by
2013-09-29 19:42:33 +0200
9965d4800
added Makefile.arm by
2013-09-29 18:55:21 +0200
4a474ea7d
changed dgemm_kernel to use fused multiply add by
2013-09-29 17:46:23 +0200
69ce737cc
modified Makefile.L3 for ARM by
2013-09-28 19:13:47 +0200
d13788d1b
common files modified for ARM by
2013-09-28 19:10:32 +0200
70411af88
initial checkin of kernel/arm by
2013-09-28 19:02:25 +0200
16eb780e1
Refs #262. Fixed compatibility issues of GNU stack markings with PathScale EKOPath(tm) Compiler Suite: Version 4.0.12.1 by
2013-09-22 09:37:59 +0800
572908899
(refs/pull/298/head)
Initial checkin of port for ARM by
2013-09-16 14:41:37 +0200
a746724e8
Added backers. by
2013-09-05 15:39:45 +0800
8cd23f206
(refs/pull/290/merge)
Merge baff5d6ba6 into 3f7b0cd994 by
2013-08-28 09:36:56 -0700
3f7b0cd99
Merge pull request #290 from larsmans/missing-threshold by
2013-08-28 17:20:16 +0200
cc6db2ecf
Merge pull request #291 from larsmans/fix-makefile-prefix by
2013-08-28 09:26:16 -0700
01e4c1354
(refs/pull/291/merge)
Merge a29e6592da into 3175be4b3d by
2013-08-28 09:25:33 -0700
3175be4b3
Merge pull request #289 from larsmans/no-noconst by
2013-08-28 09:25:23 -0700
a29e6592d
(refs/pull/291/head)
fix default prefix handling in makefiles by
2013-08-28 17:39:54 +0200
baff5d6ba
(refs/pull/290/head)
check if GEMM_MULTITHREAD_THRESHOLD defined in gemm.c by
2013-08-28 17:20:16 +0200
342af7870
(refs/pull/289/merge)
Merge 212463dce9 into 037bd82bef by
2013-08-28 07:55:46 -0700
212463dce
(refs/pull/289/head)
get rid of the generated cblas_noconst.h file by
2013-08-28 16:52:24 +0200
037bd82be
Merge pull request #288 from sebastien-villemot/develop by
2013-08-28 06:26:37 -0700
bdff0a950
(refs/pull/288/merge)
Merge eae4cfa3f6 into fe98de2f68 by
2013-08-28 05:34:53 -0700
eae4cfa3f
(refs/pull/288/head)
Avoid failure on qemu guests declaring an Athlon CPU without 3dnow! by
2013-08-28 14:27:59 +0200
6c4a7d082
Import AMD Piledriver DGEMM kernel generated by AUGEM. So far, this kernel doesn't deal with edge. by
2013-08-25 10:16:01 -0300
fe98de2f6
Merge branch 'bulldozer' into develop by
2013-08-24 11:46:18 -0300
db389b591
Refs #281. Detect __CYGWIN__ macro for Cygwin x86_64. by
2013-08-24 13:09:49 +0800
52f587db7
Refs #281. Detect _WIN32 macro for Windows API. by
2013-08-24 01:10:02 +0800
067e8417f
removed unnessesary instructions from zgemm_kernel_2x2_bulldozer.S by
2013-08-17 06:46:17 +0200
a82da3d06
removed unnessesary instructions by
2013-08-16 20:23:34 +0200
1569bf14f
Refs #282. Fixed zgemv_n typo bug on Win64. by
2013-08-23 16:27:17 +0800
df554aebd
Merge pull request #280 from ViralBShah/develop by
2013-08-21 08:21:51 -0700
fe4ca7e03
(refs/pull/280/merge)
Merge eae6920f2d into c92ae012a6 by
2013-08-21 06:45:44 -0700
eae6920f2
(refs/pull/280/head)
Patch LAPACK XLASD4.f as discussed in JuliaLang/julia#2340 by
2013-08-21 19:14:07 +0530
c92ae012a
Refs #279. Provide ONLY_CBLAS flag. If you only need CBLAS without a fortran compiler, please try make ONLY_CBLAS=1. by
2013-08-21 00:03:25 +0800
f51a849d9
Merge pull request #278 from wernsaar/haswell by
2013-08-17 08:24:37 -0700
333c1b843
(refs/pull/277/head)
removed unnessesary instructions from zgemm_kernel_2x2_bulldozer.S by
2013-08-17 06:46:17 +0200
fa3f1cd12
removed unnessesary instructions by
2013-08-16 20:23:34 +0200
035605ffe
(refs/pull/278/merge)
Merge 44ef70420c into 2638370844 by
2013-08-16 10:32:31 -0700
44ef70420
(refs/pull/278/head)
added cgemm_kernel_8x2_haswell.S by
2013-08-16 18:54:56 +0200
d488b1b1a
added zgemm_kernel_4x2_haswell.S by
2013-08-16 10:29:47 +0200
4070d9a12
added dgemm_kernel_16x2_haswell.S by
2013-08-15 19:17:20 +0200
0b90c0ec6
added sgemm_kernel_16x4_haswell.S by
2013-08-15 18:46:14 +0200
2b8ab8f55
sgemm_kernel_16x4_haswell.S minor changes by
2013-08-14 01:44:41 +0200
1cb9579cd
added zgemm_kernel_4x2_haswell.S and fixed a bug in sgemm_kernel_16x4_haswell.S by
2013-08-14 01:23:15 +0200
263837084
Init code base for Intel Haswell. by
2013-08-13 00:54:59 +0800
89637f87c
added sgemm- and dgemm-kernel for HASWELL processor by
2013-08-12 18:04:10 +0200
c0b1e41be
Merge branch 'bulldozer' into develop by
2013-08-12 23:22:10 +0800
49faee1a5
Fixed #276. Merge branch 'wernsaar-develop' into bulldozer by
2013-08-09 10:49:44 +0800
c0159d44a
Merge branch 'develop' of https://github.com/wernsaar/OpenBLAS into wernsaar-develop by
2013-08-09 10:48:46 +0800
b3220e63e
(refs/pull/276/merge)
Merge c17a850c1c into 79ba52115d by
2013-08-08 09:03:35 -0700
c17a850c1
(refs/pull/276/head)
modified KERNEL.BULLDOZER by
2013-08-08 17:49:30 +0200
099853fff
added dtrsm_kernel_RN_8x2_bulldozer.S by
2013-08-08 07:14:08 +0200
44d23881b
dtrsm_kernel_LT_8x2_bulldozer.S performance optimization by
2013-08-05 11:27:16 +0200
2905042c6
Refs #270 #268. Merge branch 'wernsaar-develop' into bulldozer by
2013-08-05 16:17:15 +0800
32fb6b9bb
Merge branch 'develop' of https://github.com/wernsaar/OpenBLAS into wernsaar-develop by
2013-08-05 16:09:47 +0800
7871dc133
(refs/pull/270/merge)
Merge aaeb8eaecd into 79ba52115d by
2013-08-05 01:09:28 -0700
673e453b3
Enable bulldozer kernels. by
2013-08-05 16:07:54 +0800
143cca4dd
Merge branch 'develop' into bulldozer by
2013-08-05 15:51:53 +0800
aaeb8eaec
(refs/pull/270/head)
modified dtrsm_kernel_LT_8x2_bulldozer.S by
2013-08-04 12:16:12 +0200
8aeec32ea
modified dtrsm_kernel_LT_8x2_bulldozer.S by
2013-08-04 10:15:33 +0200
87fc9de57
added dtrsm_kernel_LT_8x2_bulldozer.S by
2013-08-04 09:54:40 +0200